Control method, device and automatic winding system for automatic winding of lithium battery pole pieces
By image processing and edge fitting of the lithium battery electrode material, the cutting offset is calculated to achieve accurate labeling, the problems of high maintenance costs and low applicability in the existing technology are solved, and production efficiency and labeling are improved Accuracy.
Patent Information
- Application Number
- CN202411280918.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-13
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-09-13
AI Technical Summary
The existing lithium battery pole plate labeling and winding process has high maintenance costs, low applicability and insufficient flexibility, and requires multiple debugging and precise parameter calculations, resulting in low production efficiency.
By filtering and edge enhancement of the image of the cut coil, smooth continuous edge pixel coordinates are extracted, linear data is fitted using the least squares method, the cutting offset is calculated, and the actual labeling distance is calculated based on the preset labeling distance. Control the signal to the labeling mechanism to achieve accurate labeling.
It improves the accuracy and consistency of labeling, reduces maintenance costs, enhances the applicability and flexibility of the system, and improves production efficiency and quality control efficiency.
Smart Images

Figure CN118790790B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of production and manufacturing technology of lithium battery pole pieces, and in particular to a control method and device for automatic winding of lithium battery pole pieces and an automatic winding system. Background Art
[0002] In the production and manufacturing process of lithium battery pole pieces, it is generally necessary to continuously process the coiled material, i.e., the lithium battery pole piece, unwind it using a conveyor roller, and gradually wind the coiled material into a large roll using a winder, so as to achieve efficient and low-loss storage and transportation of the coiled material of the lithium battery pole piece. When the winder winds the coiled material to a full roll, it is generally necessary to manually change the roll, specifically: stop the production equipment, and fix the coiled material at the place to be cut by a fixing mechanism, then manually cut the coiled material and manually rotate to collect the remaining part, and finally stick tape to the cut place to fix the coiled material to complete the winding. For specific winding examples and finished products, please refer to Figure 1 , and further unwind the coiled material. At the same time, for the cut part of the coil, the new winding process continues. However, the efficiency of labeling, winding and quality control by manual methods is low, and it may even require multiple people to work together to change the coiled material, which is easy to delay production efficiency.
[0003] Based on this, the existing technology uses an automatic cutting device and a labeling mechanism to achieve automatic labeling and winding, specifically: the coil of lithium battery pole pieces that are nearly fully rolled is cut by the cutting mechanism, and the coil is slowly rotated to gradually reel in the tail (cut part) of the coil. Then, according to a preset rotation distance / time, the labeling mechanism is controlled by a programmable logic controller (PLC) to label the cut edge of the coil to complete the winding, and the PLC is used to control the mechanical arm, the guide rail system or the loading equipment to change the coil of lithium battery pole pieces that have been rolled.
[0004] However, in order to achieve labeling and winding by a preset rotation distance or time, equipment engineers need to accurately calculate various parameters when installing the equipment, such as the rotation distance, time and cutting position of the coil, which leads to increased installation and maintenance costs. In addition, when cutting the lithium battery pole pieces and winding them, the change in their tension affects the way the pole pieces are stacked on the reel, which in turn leads to large errors in the preset rotation distance or time. Therefore, equipment engineers need to accurately calculate various parameters when installing the equipment, including but not limited to the thickness, tension, speed and specific distance or time of the coil when rotating, which greatly increases the complexity of installation, commissioning and maintenance. In addition, when the equipment is affected by environmental factors, changes in coil characteristics, emergencies or requires emergency adjustments, it is usually necessary to rely on the manufacturer for re-debugging and maintenance, thereby delaying the production of the production line. Accordingly, the prior art has the problems of high maintenance cost, low applicability and insufficient flexibility in the labeling and winding process. Summary of the invention
[0005] Based on this, the object of the present invention is to provide a control method for automatic winding of lithium battery pole pieces.
[0006] A control method for automatically winding a lithium battery pole piece comprises the following steps:
[0007] S1: Filter and edge enhance the image of the cut coil in sequence to obtain the gradient amplitude and gradient direction of all pixels, and refine the gradient amplitude and gradient direction of all pixels through non-maximum suppression and double thresholding to obtain smooth and continuous edge pixel coordinates;
[0008] S2: linear fitting is performed on the smooth continuous edge pixel coordinates by the least square method to obtain linear data;
[0009] S3: Calculate the distance between the center of the filtered image and the straight line data to obtain the cutting offset;
[0010] S4: Calculate the actual labeling distance according to the cutting offset and the preset labeling distance, and send a control signal to the labeling mechanism according to the actual labeling distance, so that the labeling mechanism labels the cut edge of the coil to complete the winding.
[0011] The control method for automatic winding of lithium battery pole pieces described in the present invention obtains edge data of the cut portion of the coil by edge extraction and straight line fitting, calculates the cutting offset based on the edge data, combines the cutting offset with the preset labeling distance, and obtains the actual labeling distance. The labeling position is automatically adjusted through the precise actual labeling distance, thereby improving the accuracy and consistency of labeling, so as to solve the problems of high maintenance cost, low applicability and insufficient flexibility caused by the need for multiple debugging, testing and precise calculation of preset parameters in the prior art.
[0012] Furthermore, the step S1 includes the following sub-steps:
[0013] S11: Performing Gaussian filtering on the image of the cut coil to obtain a filtered image;
[0014] S12: Use the Sobel operator to calculate the gradient magnitude and gradient direction of the filtered image to obtain the gradient magnitude and gradient direction of all pixels; where the gradient direction of all pixels is The specific calculation is as follows:
[0015]
[0016] In the formula, and They are horizontal gradient image and vertical gradient image respectively, which are obtained by convolution calculation of filter images using Sobel operator;
[0017] S13: Processing the edge contour image by non-maximum suppression to obtain edge pixel coordinates;
[0018] The non-maximum suppression is processed as follows:
[0019]
[0020] In the formula, is the gradient amplitude corresponding to the pixel point of the image, and Respectively expressed as:
[0021]
[0022]
[0023] In the formula, , and , According to the pixel coordinates of the image The corresponding gradient direction Sure;
[0024] S14: Processing edge pixel coordinates through double thresholds to obtain smooth and continuous edge pixel coordinates.
[0025] Based on this, the present invention fully denoises the edge data by using Gaussian filtering, non-maximum suppression and double thresholds to ensure that the edge data is the actual edge in the image, rather than noise or pseudo-edge, thereby improving the accuracy of edge extraction.
[0026] In one embodiment, step S4 includes the following sub-steps:
[0027] The actual labeling distance is obtained by adding the preset labeling distance and the cutting offset;
[0028] Monitor the winding distance of the current winding machine rotation, determine whether the winding distance is equal to the actual labeling distance, if so, send a control signal to the labeling mechanism; if not, continue to monitor the winding distance of the current winding machine rotation.
[0029] By real-time monitoring of the winding machine's rotation distance, it is ensured that the labeling mechanism can accurately label when the coil reaches the actual labeling distance, so as to avoid labeling deviations caused by errors, thereby ensuring the quality and consistency of the winding process.
[0030] A control device for automatic winding of lithium battery pole pieces, comprising an image edge extraction unit, an edge straight line processing unit, an offset calculation unit and a control signal sending unit;
[0031] The image edge extraction unit is used to filter and enhance the edge of the image of the cut coil in sequence, obtain the gradient amplitude and gradient direction of all pixel points, and refine the gradient amplitude and gradient direction of all pixel points through non-maximum suppression and double thresholds to obtain smooth and continuous edge pixel coordinates;
[0032] The edge straight line processing unit is used to perform straight line fitting on the smooth continuous edge pixel coordinates by the least square method to obtain straight line data;
[0033] The offset calculation unit is used to calculate the distance between the center of the filtered image and the straight line data to obtain the cutting offset;
[0034] The control signal sending unit is used to calculate the actual labeling distance according to the cutting offset and the preset labeling distance, and send a control signal to the labeling mechanism according to the actual labeling distance.
[0035] Furthermore, the image edge extraction unit includes an image filtering module, an edge strength calculation module, an edge contour processing module and an edge smoothing processing module;
[0036] The image filtering module is used to perform Gaussian filtering on the image of the cut coil to obtain a filtered image;
[0037] The edge strength calculation module is used to calculate the gradient amplitude and gradient direction of the filtered image using the Sobel operator to obtain the gradient amplitude and gradient direction of all pixel points; wherein the gradient direction of all pixel points The specific calculation is as follows:
[0038]
[0039] In the formula, and They are horizontal gradient image and vertical gradient image respectively, which are obtained by convolution calculation of filter images using Sobel operator;
[0040] The edge contour processing module is used to process the edge contour image by non-maximum suppression to obtain edge pixel coordinates; wherein the non-maximum suppression is processed as follows:
[0041]
[0042] In the formula, is the gradient amplitude corresponding to the pixel point of the image, and Respectively expressed as:
[0043]
[0044]
[0045] In the formula, , and , According to the pixel coordinates of the image The corresponding gradient direction Sure;
[0046] The edge smoothing processing module is used to process edge pixel coordinates through double thresholds to obtain smooth and continuous edge pixel coordinates.
[0047] Furthermore, the control signal sending unit is used to perform the following steps:
[0048] The actual labeling distance is obtained by adding the preset labeling distance and the cutting offset;
[0049] Monitor the winding distance of the current winding machine rotation, determine whether the winding distance is equal to the actual labeling distance, if so, send a control signal to the labeling mechanism; if not, continue to monitor the winding distance of the current winding machine rotation.
[0050] An automatic winding system, comprising a winding machine, a cutting mechanism, a camera device, a labeling mechanism, and a control device for automatic winding of lithium battery pole pieces electrically and / or communicatively connected to the winding machine, the camera device and the labeling mechanism;
[0051] The winding machine comprises a rotating motor and a winding rod, and the winding rod is rotated by the rotating motor to wind the coil;
[0052] The cutting mechanism is used to cut the coiled material that reaches a preset length and / or weight;
[0053] The camera device is used to capture the image of the cut coil on the coiler and transmit the image to the control device for automatic coiling of the lithium battery pole piece;
[0054] The labeling mechanism is used to label the cut edge of the coil in the coiler according to the control signal given by the control device for automatic coiling of the lithium battery pole piece to complete the coiling;
[0055] Among them, the control device for automatic winding of lithium battery pole pieces is the control device for automatic winding of lithium battery pole pieces described above.
[0056] Furthermore, after the camera device takes pictures and before transmitting the image of the cut coil to the control device for automatically winding the lithium battery pole piece, it also includes the steps of: using a computer vision algorithm to detect the knife edge of the image of the cut coil and determine whether there is a knife edge: if so, transmitting the image of the cut coil to the control device for automatically winding the lithium battery pole piece and stopping shooting; if not, the camera device continues to shoot and executes this step.
[0057] Furthermore, after the knife edge detection shows the existence of a knife edge, the method further includes the steps of: locating the knife edge of the image of the cut material coil to obtain relevant pixel coordinate position information of the knife edge area; and generating an image of the key area by combining the relevant pixel coordinate position information with the image of the cut material coil, and using the image as the image of the cut material coil.
[0058] Furthermore, the automatic winding system also includes an automatic guided vehicle; the automatic guided vehicle is used to pull the coiled material after the winding machine completes winding away from the current area, thereby realizing fully automatic roll changing and unloading.
[0059] For better understanding and implementation, the present invention is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0060] Figure 1 It is an example of winding of coil / substrate and a simple schematic diagram of the final product;
[0061] Figure 2 It is a simplified structural schematic diagram of the automatic winding system of the present invention;
[0062] Figure 3 It is a simple structural diagram of the automatic winding system after the coil is cut;
[0063] Figure 4 This is a schematic diagram of the structure of the control device for automatically winding up the lithium battery pole piece according to the present invention;
[0064] Figure 5 This is a simplified flow chart of the control method for automatic winding of lithium battery pole pieces according to the present invention;
[0065] Figure 6 It is a schematic diagram of the simulation application of edge extraction and labeling area of the present invention;
[0066] Figure 7 It is a schematic diagram of the knife edge after cutting;
[0067] Figure 8 A simplified example diagram of the cutting offset and preset labeling distance. DETAILED DESCRIPTION
[0068] In order to solve the problems of high maintenance cost, low applicability and insufficient flexibility in the labeling and winding process of the prior art, the present invention obtains the image of the coil after cutting, extracts the edge data of the cut part of the image, and fits the straight line data of the cut edge, calculates the offset of the cutting based on the straight line data, calculates the precise and actual labeling distance based on the offset of the cutting, rotates and rewinds the coil based on the quasi-labeling distance, and labels the coil to complete the winding operation. Accordingly, the present invention introduces an edge extraction algorithm to ensure the accuracy of obtaining the cutting and labeling positions, reduce maintenance costs, improve the applicability and flexibility of the system, and thus improve production efficiency and quality control efficiency.
[0069] Based on the above design, the present invention proposes a control method for automatic winding of lithium battery pole pieces, and based on this method, proposes a control device for automatic winding of lithium battery pole pieces.
[0070] Please also see Figure 2 and Figure 3 , Figure 2 This is a simple structural diagram of the automatic winding system. Figure 3 It is a simple structural diagram of the automatic winding system after the coil is cut.
[0071] An automatic winding system includes: a winding machine 101, a cutting mechanism 103, a camera device 105, a labeling mechanism 107, and a control device 110 for automatic winding of lithium battery pole pieces electrically and / or communicatively connected to the winding machine, the camera device and the labeling mechanism.
[0072] The winding machine 101 is arranged at the tail of the coil and is used to wind the coil. Specifically, it includes a rotating motor and a winding rod. The rotating motor rotates the winding rod to wind the coil. According to different production requirements or process standards, the winding machine may also include a fixing mechanism and other structures. The present invention does not specifically limit the selection of the winding machine.
[0073] The cutting mechanism 103 is arranged on the path of the coiled material, and is used to cut the coiled material reaching a preset length and / or weight in the coiler during the winding process, so as to ensure smooth winding of the coiled material.
[0074] The camera device 105 is used to capture the image of the cut coil on the winder, and transmit the image of the cut coil to the control device 110 for automatic winding of the lithium battery pole piece. Specifically, after the cutting mechanism 103 performs cutting, the programmable logic controller (PLC) monitors the rotation distance of the winder 101. When the rotation distance reaches the preset distance, the camera device is called to shoot to accurately obtain the image of the cut coil. The preset distance is the approximate distance between the cutting mechanism and the camera device. For example, if the mechanical distance from the cutting device to the camera device is 5 meters, the preset distance can be selected from a distance range of 4.5 meters to 5.5 meters. Among them, the camera device can be selected as a charge-coupled device (CCD).
[0075] The present invention monitors the rotation distance of the winder to reach a preset distance to ensure that the camera device 105 only starts shooting when the blade of the coil enters the shooting range, thereby preventing real-time edge detection from being easily disturbed by light and shadow or other environmental factors, resulting in false detection.
[0076] Furthermore, after the camera device takes pictures and before transmitting the image of the cut coil to the control device for automatically winding the lithium battery pole piece, it also includes using a computer vision algorithm to detect the knife edge of the image of the cut coil and determine whether there is a knife edge: if so, the image of the cut coil is transmitted to the control device for automatically winding the lithium battery pole piece and the shooting is stopped; if not, the camera device continues to shoot and executes this step.
[0077] The computer vision algorithm can be optionally trained through a machine learning or deep learning algorithm, and assisted by relevant knife edge images, so as to obtain a corresponding binary classification computer vision algorithm. Specifically, support vector machine (SVM), multi-layer perceptron (MLP), random forest or convolutional neural network can be used to realize binary classification of knife edge detection. Accordingly, the present invention does not specifically limit the selection of its computer vision algorithm.
[0078] By using computer vision algorithms to detect the knife edge before performing edge extraction, unnecessary edge extraction operations can be reduced to reduce unnecessary interference, thereby reducing the accuracy of its preset distance debugging, thereby reducing maintenance costs and improving flexibility.
[0079] Furthermore, after the presence of a knife edge is detected, the knife edge is located on the image of the cut material coil, and relevant pixel coordinate position information of the knife edge area is obtained; and by combining the relevant pixel coordinate position information with the image of the cut material coil, an image of the key area is generated, and the image is transmitted as the image of the cut material coil to the control device for automatically winding the lithium battery pole piece.
[0080] The image of the cut coiled material in the key area enables the automatic winding control device to process the relevant pixel coordinate position information first.
[0081] Accordingly, the control device for automatically winding up the lithium battery pole piece is made to focus on the knife edge area to improve the accuracy and efficiency of knife edge positioning.
[0082] The labeling mechanism 107 is used to attach the tape to the cut edge of the coil on the winding machine according to a given control signal to complete the winding.
[0083] The control device 110 for automatically winding the lithium battery pole piece is used to obtain the image of the cut coil taken by the camera device 105, calculate the cutting edge data in the image, and calculate the appropriate conditions based on the cutting edge data, and send a control signal to the labeling mechanism 107 based on the conditions. The control device can be optionally implemented by a programmable logic controller (PLC).
[0084] Furthermore, the automatic winding system also includes an automated guided vehicle (AGV), which is used to pull the lithium battery pole pieces after labeling and winding in the winding machine away from the current area, thereby realizing fully automatic roll changing and unloading.
[0085] Please also see Figure 4 and Figure 5 , Figure 4 This is a schematic diagram of the structure of the control device for automatically winding up the lithium battery pole piece of the present invention. Figure 5 The present invention is a simplified flow chart of the control method for automatically winding up the lithium battery pole piece.
[0086] The control device for automatically rolling up a lithium battery pole piece comprises an image edge extraction unit 1, an edge straight line processing unit 2, an offset calculation unit 3 and a control signal sending unit 4.
[0087] The image edge extraction unit 1 is used to execute step S1: filtering and edge enhancing the image of the cut coil in sequence to obtain the gradient amplitude and gradient direction of all pixel points, and refining the gradient amplitude and gradient direction of all pixel points through non-maximum suppression and double thresholds to obtain smooth and continuous edge pixel coordinates.
[0088] Specifically, the image edge extraction unit 1 includes an image filtering module 11 , an edge strength calculation module 12 , an edge contour processing module 13 and an edge smoothing processing module 14 .
[0089] The image filtering module 11 is used to execute step S11: perform Gaussian filtering on the image of the cut coil material to obtain a filtered image.
[0090] Specifically, the Gaussian filter performs a convolution operation on the image based on the Gaussian kernel to remove noise in the image, thereby obtaining a smooth image, that is, a filtered image. The specific expression is as follows:
[0091]
[0092] in, Represents the pixel coordinates in the image after Gaussian filter convolution The value of is the standard deviation, which is used to determine the width of the Gaussian kernel; is a normalization constant used to ensure that the sum of all elements of the Gaussian kernel is 1.
[0093] Gaussian filtering can significantly reduce high-frequency noise in the image, making the edges of the image smoother and clearer for edge extraction in subsequent steps.
[0094] The edge strength calculation module 12 is used to execute step S12: using the Sobel operator to calculate the gradient magnitude and gradient direction of the filtered image to obtain the gradient magnitude and gradient direction of all pixel points.
[0095] Specifically, the horizontal gradient image of the filtered image is calculated using the horizontal convolution kernel and the vertical convolution kernel. and vertical gradient image , the horizontal convolution kernel is expressed as follows:
[0096]
[0097] The vertical convolution kernel is expressed as follows:
[0098]
[0099] Next, the horizontal gradient image and vertical gradient image Perform amplitude calculation to obtain the gradient amplitude, which is specifically expressed as follows:
[0100]
[0101] in, The gradient amplitude of each pixel in the edge image is used to reflect the strength of the edge.
[0102] Finally, the gradient direction of each pixel of the filtered image is calculated by trigonometric functions, which is specifically expressed as follows:
[0103]
[0104] in, Represents the gradient direction of each pixel.
[0105] The horizontal and vertical gradients of the image are calculated by the convolution kernels in the horizontal and vertical directions of the Sobel operator, respectively, so as to obtain the gradient magnitude and direction of each pixel. The gradient magnitude reflects the intensity of the edge pixel, making the edge more prominent and clear, while the gradient direction indicates the direction of the edge. Accordingly, the gradient magnitude and direction are used to characterize the visibility of the image edge.
[0106] The edge contour processing module 13 is used to execute step S13: performing non-maximum suppression processing on the gradient amplitudes and gradient directions of all pixel points to obtain edge pixel coordinates.
[0107] Specifically, for the gradient amplitude of a pixel, the gradient amplitudes of adjacent pixels are compared along its gradient direction, and it is determined whether the gradient amplitude of the pixel is a local maximum: if so, the pixel is set to 0; if not, the gradient amplitude of the pixel is retained; the edge pixel coordinates are obtained, which are specifically expressed as follows:
[0108]
[0109] in, is the pixel coordinate of the image The corresponding gradient amplitude is and Respectively expressed as:
[0110]
[0111]
[0112] In the formula, , and , According to pixel coordinates The corresponding gradient direction Sure.
[0113] Through non-maximum suppression (NMS), the edge pixel coordinates are further refined to retain only the pixels with the largest gradient amplitude in the edge data, thereby removing the noise of non-edge points and obtaining a clearer, more accurate and detailed edge contour.
[0114] The edge smoothing processing module 14 is used to execute step S14: processing edge pixel coordinates through double thresholds to obtain smooth and continuous edge pixel coordinates.
[0115] Specifically, two thresholds are set, namely, high threshold and low threshold ,in, ;
[0116] Filter edge pixel coordinates through high thresholds to locate strong edge pixel coordinates, specifically:
[0117] If exists ,but as strong edge pixel coordinates;
[0118] in, is the pixel coordinate in the image The corresponding gradient magnitude.
[0119] Next, the strong edge pixel coordinates are filtered by a low threshold to locate the weak edge pixel coordinates, specifically:
[0120] If exists and is the neighborhood pixel of the strong edge pixel, then As weak edge pixel coordinates.
[0121] Finally, the strong edge pixel coordinates and the weak edge pixel coordinates are connected to obtain smooth and continuous edge pixel coordinates, which can be specifically expressed as:
[0122]
[0123] in, are the pixel coordinates of all smooth continuous edges of the coil cutting contour, The total number of pixel coordinates for smoothing continuous edge pixel coordinates.
[0124] Accordingly, a high threshold is used to refine strong edges, and a low threshold is used to extract weak edges, and the two are connected to ensure the continuity and integrity of the edges, thereby making the edge contour smoother and smaller.
[0125] Due to the differences in the substrate, production environment, and parameters of the camera equipment of lithium battery pole pieces, the effects of strong and weak edge extraction are different. Therefore, the high threshold and the low threshold need to be set by the user in a targeted manner. The present invention does not specifically limit the setting of the dual threshold.
[0126] Based on this, the present invention fully denoises the edge data by using Gaussian filtering, non-maximum suppression and double thresholds to ensure that the edge data is the actual edge in the image, rather than noise or pseudo-edge, thereby improving the accuracy of edge extraction.
[0127] The edge line processing unit 2 is used to execute step S2: performing straight line fitting on the smooth continuous edge pixel coordinates by the least square method to obtain straight line data.
[0128] Specifically, suppose the equation of the straight line to be fitted is ;
[0129] For all smooth continuous edge pixel coordinates , then the objective function of the least squares method (minimizing the error) is:
[0130]
[0131] Then, and To perform partial derivative:
[0132]
[0133]
[0134] And order and The partial derivative of is 0, then we get a system of linear equations in two variables:
[0135]
[0136] Solving the above equations, we can get the straight line fitting parameters:
[0137]
[0138] Based on this, the straight line fitting parameters are substituted into the straight line equation to be fitted to obtain the straight line data.
[0139] Due to the brittle nature of the active materials coated in the lithium battery pole pieces, when the stress of cutting is concentrated, uneven fractures or delamination are prone to occur. In addition, when the lithium battery pole piece is in motion and is directly cut straight, it is also easy to cause edge tearing. Accordingly, the present invention utilizes a serrated blade to ensure that force can be evenly applied and the lithium battery pole piece can be easily penetrated each time the automatic cutting is performed, thereby reducing the stress concentration generated during cutting, and further reducing the occurrence of the above situation. In addition, since there are interference points in the pixel coordinate points after edge extraction, the influence of the interference points on the cutting line can be removed after the least squares method is used to fit the straight line, thereby improving the accuracy. For details, please refer to Figure 7 , Figure 7 This is a schematic diagram of an example of a knife edge after cutting, where the more obvious lines are the lines after edge extraction.
[0140] Based on this, in order to accurately locate the edge position of the sawtooth knife edge during the edge extraction process, it is necessary to further perform straight line fitting on the edge pixels to smooth and simplify the jagged edge information, thereby providing more accurate edge representation and positioning results.
[0141] The offset calculation unit 3 is used to execute step S3: calculate the distance between the center of the filtered image and the straight line data to obtain the cutting offset.
[0142] Specifically, the difference between the center position of the filtered image and the center coordinates of the straight line data is calculated to obtain the cutting offset, which can be specifically expressed as:
[0143] Assume the size of the filtered image is , then the coordinates of the center point are :
[0144]
[0145]
[0146] Calculate straight line data To center coordinate point Distance:
[0147]
[0148] Among them, it means Cutting offset, and Line fit parameters.
[0149] By characterizing the cutting offset as the distance between the center of the image extraction and the center point of the edge, the labeling position can be accurately located, reducing the need for manual adjustment and calibration, thereby improving production efficiency and product quality. In addition, accurate positioning of the cutting position can also ensure the integrity of the edge, avoiding damage or incompleteness of the edge during transportation due to deviation in the labeling position.
[0150] See also Figure 8 , Figure 8 A simplified example diagram of the cutting offset and preset labeling distance.
[0151] The control signal sending unit 4 is used to execute step S4: calculate the actual labeling distance according to the cutting offset and the preset labeling distance, and send a control signal to the labeling mechanism according to the actual labeling distance, so that the labeling mechanism labels the cut edge of the coil to complete the winding.
[0152] Specifically, the preset labeling distance is the distance between the camera device and the labeling mechanism, and the actual labeling distance is obtained by adding the preset labeling distance to the cutting offset.
[0153] Monitor the winding distance of the current winding machine rotation, determine whether the winding distance is equal to the actual labeling distance, if so, send a control signal to the labeling mechanism; if not, continue to monitor the winding distance of the current winding machine rotation.
[0154] The present invention monitors the rotation distance of the winder and ensures that the labeling mechanism can accurately label when the coil reaches the actual labeling distance in a countdown manner, so as to avoid labeling deviation caused by errors, thereby ensuring the quality and consistency of the winding process.
[0155] Compared with the prior art, the present invention extracts the cutting edge through Gaussian filtering, non-maximum suppression and double thresholds, and refines the edge data to obtain a finer contour, thereby achieving high-precision edge extraction. In addition, in order to ensure the accuracy of the jagged edge data, the present invention simplifies and smoothes the jagged edge information through straight line fitting, and accurately locates the labeling position according to the calculated cutting offset, thereby achieving high-precision cutting edge positioning labeling and winding, and its labeling accuracy is , which significantly improves the applicability and flexibility of automatic winding. Furthermore, by combining with an automatic guided vehicle, the automatically wound coil can be automatically changed and unloaded, thus realizing a fully automated production process of winding, changing and unloading.
[0156] Based on the same inventive concept, the present application also provides an electronic device, which may be a terminal device such as a server, a desktop computing device or a mobile computing device (e.g., a laptop computing device, a handheld computing device, a tablet computer, a netbook, etc.). The device includes one or more processors and a memory, wherein the processor is used to execute a program to implement the control method for automatic winding of a lithium battery pole piece according to an embodiment of the present invention; and the memory is used to store a computer program executable by the processor.
[0157] Based on the same inventive concept, the present application also provides a computer-readable storage medium, corresponding to the embodiment of the control method for automatic winding of lithium battery pole pieces mentioned above, and the computer-readable storage medium stores a computer program thereon, and when the program is executed by the processor, the steps of the control method for automatic winding of lithium battery pole pieces recorded in any of the above embodiments are implemented.
[0158] The present application may take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-usable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.
[0159] The above-mentioned embodiments only express several implementation methods of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, and the present invention is also intended to include these modifications and modifications.
Claims
1. A control method for automatic winding of lithium battery pole pieces, characterized in that: The following steps are involved: S1: filtering and edge enhancing the image of the cut coil in sequence to obtain the gradient amplitude and gradient direction of all pixels, and refining the gradient amplitude and gradient direction of all pixels by non-maximum suppression and double thresholding to obtain smooth continuous edge pixel coordinates; wherein, step S1 includes the following sub-steps: S11: Performing Gaussian filtering on the image of the cut coil to obtain a filtered image; S12: Use the Sobel operator to calculate the gradient magnitude and gradient direction of the filtered image to obtain the gradient magnitude and gradient direction of all pixels; where the gradient direction of all pixels is The specific calculation is as follows: In the formula, and They are horizontal gradient image and vertical gradient image respectively, which are obtained by convolution calculation of filter images using Sobel operator; S13: Processing the edge contour image by non-maximum suppression to obtain edge pixel coordinates; The non-maximum suppression is processed as follows: In the formula, is the gradient amplitude corresponding to the pixel point of the image, and Respectively expressed as: In the formula, , and , According to the pixel coordinates of the image The corresponding gradient direction Sure; The image of the cut coil is obtained by taking a picture with a camera device; S14: Processing edge pixel coordinates through double thresholds to obtain smooth and continuous edge pixel coordinates; S2: Linear fitting is performed on the smooth continuous edge pixel coordinates by the least square method to obtain straight line data; S3: Calculate the distance between the center of the filtered image and the straight line data to obtain the cutting offset; S4: Calculate the actual labeling distance according to the cutting offset and the preset labeling distance, and send a control signal to the labeling mechanism according to the actual labeling distance, so that the labeling mechanism labels the cut edge of the coil to complete the winding; wherein, the step S4 includes the following sub-steps: The actual labeling distance is obtained by adding the preset labeling distance and the cutting offset; Monitor the winding distance of the current winding machine rotation, determine whether the winding distance is equal to the actual labeling distance, if so, send a control signal to the labeling mechanism; if not, continue to monitor the winding distance of the current winding machine rotation; Wherein, the preset labeling distance is the distance between the camera device and the labeling mechanism.
2. A control device for automatically winding a lithium battery pole piece, characterized in that: It includes an image edge extraction unit, an edge straight line processing unit, an offset calculation unit and a control signal sending unit; The image edge extraction unit is used to filter and enhance the edge of the image of the cut coil in sequence, obtain the gradient amplitude and gradient direction of all pixel points, and refine the gradient amplitude and gradient direction of all pixel points through non-maximum suppression and double thresholds to obtain smooth and continuous edge pixel coordinates; Wherein, the image edge extraction unit includes an image filtering module, an edge strength calculation module, an edge contour processing module and an edge smoothing processing module; The image filtering module is used to perform Gaussian filtering on the image of the cut coil to obtain a filtered image; The edge strength calculation module is used to calculate the gradient amplitude and gradient direction of the filtered image using the Sobel operator to obtain the gradient amplitude and gradient direction of all pixel points; wherein the gradient direction of all pixel points The specific calculation is as follows: In the formula, and They are horizontal gradient image and vertical gradient image respectively, which are obtained by convolution calculation of filter images using Sobel operator; The edge contour processing module is used to process the edge contour image by non-maximum suppression to obtain edge pixel coordinates; wherein the non-maximum suppression is processed as follows: In the formula, is the gradient amplitude corresponding to the pixel point of the image, and Respectively expressed as: In the formula, , and , According to the pixel coordinates of the image The corresponding gradient direction Sure; The image of the cut coil is obtained by taking a picture with a camera device; The edge smoothing processing module is used to process edge pixel coordinates through double thresholds to obtain smooth and continuous edge pixel coordinates; The edge straight line processing unit is used to perform straight line fitting on the smooth continuous edge pixel coordinates by the least square method to obtain straight line data; The offset calculation unit is used to calculate the distance between the center of the filtered image and the straight line data to obtain the cutting offset; The control signal sending unit is used to calculate the actual labeling distance according to the cutting offset and the preset labeling distance, and send a control signal to the labeling mechanism according to the actual labeling distance; wherein the control signal sending unit is specifically used to perform the following steps: The actual labeling distance is obtained by adding the preset labeling distance and the cutting offset; Monitor the winding distance of the current winding machine rotation, determine whether the winding distance is equal to the actual labeling distance, if so, send a control signal to the labeling mechanism; if not, continue to monitor the winding distance of the current winding machine rotation; Wherein, the preset labeling distance is the distance between the camera device and the labeling mechanism.
3. An automatic winding system, characterized in that: It includes a winding machine, a cutting mechanism, a camera device, a labeling mechanism, and a control device for automatically winding a lithium battery electrode electrically and / or communicatively connected to the winding machine, the camera device and the labeling mechanism; The winding machine comprises a rotating motor and a winding rod, and the winding rod is rotated by the rotating motor to wind the coil; The cutting mechanism is used to cut the coiled material that reaches a preset length and / or weight; The camera device is used to capture the image of the cut coil on the coiler and transmit the image to the control device for automatic coiling of the lithium battery pole piece; The labeling mechanism is used to label the cut edge of the coil in the coiler according to the control signal given by the control device for automatic coiling of the lithium battery pole piece to complete the coiling; Wherein, the control device for automatic winding of lithium battery pole pieces is the control device for automatic winding of lithium battery pole pieces as claimed in claim 2.
4. The automatic winding system according to claim 3, characterized in that: After the camera device takes pictures and transmits the image of the cut coil to the control device for automatically winding the lithium battery pole piece, it also includes the steps of: using a computer vision algorithm to detect the knife edge of the image of the cut coil and determine whether there is a knife edge: if so, transmit the image of the cut coil to the control device for automatically winding the lithium battery pole piece and stop shooting; if not, the camera device continues to shoot and executes this step.
5. The automatic winding system according to claim 4, characterized in that: After the knife edge detection is performed to determine the existence of a knife edge, the step further includes: positioning the knife edge on the image of the cut coil material to obtain relevant pixel coordinate position information of the knife edge area; and combining the relevant pixel coordinate position information with the image of the cut coil material to generate an image of the key area, and using the image as the image of the cut coil material.
6. The automatic winding system according to any one of claims 3 to 5, characterized in that: It also includes an automatic guided vehicle; the automatic guided vehicle is used to pull the coiled material away from the current area after the coiler completes winding.
